Process Evaluation in an Intervention to Improve Colorectal Cancer Screening

Ferreira MR, Fitzgibbon ML, Dolan NC, Davis TC, Rademaker AW, Wolf MS, Liu D, Gorby N, Schmitt BP, Bennett CL. Process Evaluation in an Intervention to Improve Colorectal Cancer Screening. Paper presented at: VA HSR&D National Meeting; 2005 Feb 1; Baltimore, MD.

Objectives:Colorectal cancer (CRC) is the third most common cancer in the US. Although CRC screening is recommended for individuals 50 years and older, screening rates are low, due to patient and provider barriers. We developed and implemented a combined patient/provider intervention, randomized by clinic, at VA Chicago. The primary aim was to increase CRC screening recommendations and completion. We conducted a process evaluation to examine the relationship between intervention participation and outcomes among patients and providers in the intervention clinic. Methods:Patients were males, 50 years and older, attending the intervention clinic, at average risk for CRC, and noncompliant with screening. The patient component included a video and brochure. Providers were physicians and nurse practitioners at the clinic. The physician component included an educational workshop and quarterly sessions with feedback on screening recommendations and adherence. Provider and patient demographics were compared using the independent sample t-test or Fisher's exact test. Recommendation and completion rates were compared between provider and patient groups. Results:Of 44 providers, 37 attended at least one intervention session. There were no differences in gender, training or race/ethnicity between providers who attended sessions and providers who did not. Of 986 patients, only 258 received the intervention. There were no age or race/ethnicity differences between patient groups; those who received the intervention had more clinic visits (3.4 vs 2.6). Providers who attended sessions recommended CRC screening more frequently than providers who did not (64% vs 54% of visits, p < 0.01). Patients of providers who attended sessions were more likely to complete screening (42% versus 29%, p < 0.05). Patients who participated in the patient intervention were more likely to receive screening recommendations (83.7% vs 74.6%, p < 0.01); however, they were as likely to complete screening. Implications:The process analysis showed that the provider component was implemented as designed. This minimal form of provider-directed intervention increased CRC screening recommendations and completion. The patient component was not fully implemented, due to organizational difficulties of delivering patient interventions in busy clinical settings. Impacts:Future studies should plan for strategies to facilitate successful implementation.